Table Detection in Spreadsheet

1. A computer-implemented method, comprising: extracting respective multiple attributes of multiple cells comprised in a spreadsheet; determining respective features of the multiple cells based on the extracted multiple attributes; dividing the multiple cells into at least one candidate region based on the features and a mapping relationship between a table in a spreadsheet and features of multiple cells comprised in the table, the mapping relationship being trained based on features of multiple cells comprised in a set of sample spreadsheets and multiple tables comprised in the set of sample spreadsheets to determine a probability that a table exists; determining at least one candidate table in the spreadsheet based on the at least one candidate region; and determining whether a potential error exists in the at least one candidate region based at least on probabilistic information describing the probability that the table exists in the at least one candidate region based on the mapping relationship.

2. The method of claim 1 , wherein extracting respective multiple attributes of multiple cells comprised in the spreadsheet comprises: for a given cell among the multiple cells, extracting multiple attributes of the given cell based on at least one of characters of data in the given cell, format of the data in the given cell, or style of the given cell.

3. The method of claim 1 , wherein determining the at least one candidate region further comprises: for a given candidate region of the at least one candidate region, adjusting a boundary of the given candidate region based on a matching degree of the given candidate region with the mapping relationship, so as to update the given candidate region.

4. The method of claim 1 , wherein the determining whether the potential error exists in the at least one region comprises: obtaining at least one real table in the spreadsheet in response to determining a potential error exists in a given candidate region; and updating the mapping relationship based on the at least one real table and features of the multiple cells in the spreadsheet.

5. The method of claim 1 , further comprising: obtaining information describing the probability that a given candidate region of at least one candidate region obtained according to the mapping relationship comprises the table, and wherein determining whether the potential error exists in the given candidate region of the at least one candidate region comprises: determining the probability that e given candidate region comprises the table based on the information; and determining the potential error exists in the given candidate region in response to the probability being below a predefined threshold probability.

6. The method of claim 1 , further comprising: obtaining information describing an additional probability that a cell in a spreadsheet is within the table, and determining whether the potential error exists in a given candidate region of the at least one candidate region comprises: determining the additional probability that the cell in the given candidate region is within the table based on the information; and determining the potential error exists in the given candidate region in response to the additional probability being below a predefined threshold probability.

7. The method of claim 1 , wherein determining whether the potential error exists in the at least one candidate region comprises: determining the potential error exists in a given candidate region of the at least one candidate region in response to the given candidate region overlapping with a further candidate region of the at least one candidate region.

8. The method of claim 1 , wherein determining whether the potential error exists in the at least one candidate region comprises: determining the potential error exists in a given candidate region of the at least one candidate region in response to an edge of the given candidate region comprising at least one of an empty row and an empty column.

9. The method of claim 1 , wherein determining whether the potential error exists in the at least one candidate region comprises: determining whether the potential error exists in a given candidate region of the at least one candidate region based on a location relationship between a non-empty cell among the multiple cells and the given candidate region.

10. A device, comprising: a processing unit; a memory coupled to the processing unit and including instructions stored thereon, the instructions, when executed by the processing unit, causing the device to perform operations comprising: extracting respective multiple attributes of multiple cells comprised in a spreadsheet; determining respective features of the multiple cells based on the extracted multiple attributes; dividing the multiple cells into at least one candidate region based on the features and a mapping relationship between a table in a spreadsheet and features of multiple cells comprised in the table the mapping relationship being trained based on features of multiple cells comprised in a set of sample spreadsheets and multiple tables comprised in the set of sample spreadsheets to determine a probability that a table exists; determining at least one candidate table in the spreadsheet based on the at least one candidate region; and determining whether a potential error exists in the at least one candidate region based at least on probabilistic information describing the probability that a table exists in the at least one candidate region based on the mapping relationship.

11. The device of claim 10 , wherein determining the at least one candidate region further comprises: for a given candidate region of the at least one candidate region, adjusting a boundary of the given candidate region based on a matching degree of the given candidate region with the mapping relationship, so as to update the given candidate region.

12. The device of claim 10 , wherein the determining whether the potential error exists in the at least one candidate region comprises: obtaining at least one real table in the spreadsheet in response to determining a potential error exists in a given candidate region; and updating the mapping relationship based on the at least one real table and features of the multiple cells in the spreadsheet.

13. A non-transitory computer readable storage medium, on which a computer program is stored, the program, when executed by at least one processor, causes the at least one processor to perform operations comprising: extracting respective multiple attributes of multiple cells comprised in a spreadsheet; determining respective features of the multiple cells based on the extracted multiple attributes; dividing the multiple cells into at least one candidate region based on the features and a mapping relationship between a table in a spreadsheet and features of multiple cells comprised in the table, the mapping relationship being trained based on features of multiple cells comprised in a set of sample spreadsheets and multiple tables comprised in the set of sample spreadsheets to determine a probability that a table exists; determining at least one candidate table in the spreadsheet based on the at least one candidate region; and determining whether a potential error exists in the at least one candidate region based at least on probabilistic information describing the probability that a table exists in the at least one candidate region based on the mapping relationship.